Forest Ecology and Management 178 (2003) 5–21
The role of climate and vegetation change in shaping past and
future fire regimes in the northwestern US and
the implications for ecosystem management
Cathy Whitlocka,*, Sarah L. Shaferb, Jennifer Marlona
a
Department of Geography, University of Oregon, Eugene, OR 97403, USA
b
U.S. Geological Survey, 200 SW 35th St., Corvallis, OR 97333, USA
Abstract
Fire is an important part of the disturbance regimes of northwestern US forests and its role in maintaining and altering forest
vegetation is evident in the paleoecological record of the region. Long-term reconstructions of Holocene fire regimes, provided
by the analysis of charcoal, pollen, and other fire proxies in a network of lake records, indicate that the Pacific Northwest and
summer-dry regions of the northern Rocky Mountains experienced their highest fire activity in the early Holocene (11,000–7000
years ago) and during the Medieval Warm Period (ca. 1000 years ago) when drought conditions were more severe than today. In
contrast, in summer-wet areas of the northern Rocky Mountains, the period of highest fire activity was registered in the last 7000
years when dry woodland vegetation developed. When synthesized across the entire northwestern US, the paleoecological
record reveals that past and present fire regimes are strongly controlled by climate changes occurring on multiple time scales.
The scarcity of fires in the 20th century in some northwestern US ecosystems may be the result of successful fire suppression
policies, but in wetter forests this absence is consistent with long-term fire regime patterns. In addition, simulations of potential
future climate and vegetation indicate that future fire conditions in some parts of the northwestern US could be more severe than
they are today. The Holocene record of periods of intensified summer drought is used to assess the nature of future fire–climate–
vegetation linkages in the region.
# 2003 Elsevier Science B.V. All rights reserved.
Keywords: Fire history; Charcoal records; Holocene climate change; Future fire conditions; Western US
1. Introduction
The role of fire as a driver of ecological change is
widely recognized among researchers and managers.
In forests of the western US much attention has
focused on trying to identify and recreate the fire
regimes that existed at the time of Euro-American
settlement and before extensive fire suppression
*
Corresponding author. Tel.: þ1-541-346-4566;
fax: þ1-541-346-2067.
E-mail address: whitlock@oregon.uoregon.edu (C. Whitlock).
efforts. The last 15 years have witnessed several large
wildfires in western coniferous forests. In 2000, for
example, over 3.5 million ha burned in western Montana, northern Idaho, and California despite over $ 1.3
billion spent on fire fighting efforts (National Interagency Fire Center, 2002). These recent fire events are
widely blamed on the build-up of fuels, in particular the
‘‘thickening’’ of the forest that has occurred in recent
decades in the absence of fire. In 2001, over $ 200
million was allocated as part of the National Fire
Plan directed at fuel mitigation and our ability to
suppress fires through such actions continues to
0378-1127/03/$ – see front matter # 2003 Elsevier Science B.V. All rights reserved.
doi:10.1016/S0378-1127(03)00051-3
6
C. Whitlock et al. / Forest Ecology and Management 178 (2003) 5–21
improve (USDA Forest Service, 2001). In the last 15
years, the western US has also experienced extreme
fire-conducive climate conditions, including droughts
that have been more severe than any previously
recorded in the instrumental meteorological record
(National Oceanic and Atmospheric Administration,
2002).
The co-occurrence of high fuel accumulation, low
fuel moisture, and fire-conducive weather patterns has
given rise to a national discussion about the causes and
proper responses to wildland fires. The debate centers
on some basic questions: to what extent is the current
fire regime caused by forest management or is it a
product of changing climate (both natural and humaninduced changes)? Irrespective of the cause, do the
environmental conditions supporting the current fire
regimes exceed the range of conditions experienced in
the past? Can intensive fuels management restore the
pre-settlement fire regimes, given the potential magnitude of environmental changes projected for the
future?
Answering these questions requires an understanding of the relationships among fire, climate, and
vegetation, especially the dynamics of those relationships prior to extensive forest management activities.
These questions also raise important management
issues about future fire conditions that may occur as
a result of potential future vegetation and climate
changes. In this paper, we describe the methods used
to study past fire, vegetation, and climate interactions
in the northwestern US, an area that includes the
Pacific Northwest and northern Rocky Mountains.
These reconstructions provide an opportunity to
examine both the pre-historical and historical range
of variability of fire occurrence on multiple time
scales. We also examine the potential future environmental changes in the region as simulated by climate
and vegetation models and discuss the implications of
these changes for future fire regimes and management
responses to fire.
2. Tools for paleoenvironmental research
2.1. Fire-history reconstructions
Fire-history information is provided primarily by
two types of data: (1) fire-scarred tree rings and stand-
age data, and (2) proxy data from sediment records
(see Agee, 1993; Patterson et al., 1987; Whitlock and
Larsen, 2002). These data sources describe fire
regimes at multiple temporal and spatial scales.
Tree-ring data and stand-age information provide
short-term reconstructions of fire events, usually spanning the last 400 years or less. Fire-scars on tree rings
register fire events that were not lethal to the tree,
whereas stand establishment dates identify the minimum age of the last stand-replacement fire. In the
northwestern US, the fire history of montane and
subalpine forests is characterized by infrequent severe
fires, as evidenced by the mosaic of stand ages in these
forests (Arno and Sneck, 1977; Romme and Despain,
1989; Agee, 1993). The dominant conifer species of
these forests, including subalpine fir (Abies lasiocarpa
(bifolia)), whitebark pine (Pinus albicaulis), and
Engelmann spruce (Picea engelmannii), are adapted
to a long fire interval (i.e., a long time between
subsequent stand-replacing fires). Other species, such
as lodgepole pine (Pinus contorta), help maintain a
stand-replacing fire regime as a result of their high
flammability and serotinous cones. Lower elevation
forests, dominated by ponderosa pine (Pinus ponderosa) and Douglas-fir (Pseudotsuga menziesii), are
maintained by frequent ground fires, which are best
measured by fire-scarred tree-rings. In between these
two extremes are fire regimes in forests that feature
mixtures of fire-resistant and fire-sensitive species,
which result in a complex array of stand ages and
species composition derived from both crown and
ground fires. Most forested regions of the western
US experience this mid-severity fire regime.
The second method of fire-history reconstruction,
and the primary focus of this paper, involves the
examination of paleoenvironmental data preserved
in sediments, with most work focusing on lake sediments (although see Meyer and Pierce, 2003 for a
discussion of fire records from alluvial fan and colluvial sediments and Gavin, 2000 for fire records from
soils). Lake sediments serve as repositories of proxy
data recording past environmental conditions within a
watershed. Particulate charcoal, such as charred wood,
bark, and leaves, is introduced into a lake through
airborne fallout and delivery by streams and slopewash. Intervals in sediment cores for which charcoal
abundance exceeds a prescribed background level are
considered evidence of a fire episode. The time series
C. Whitlock et al. / Forest Ecology and Management 178 (2003) 5–21
of charcoal peaks from lake-sediment records provide
information on long-term variations in fire frequency.
The tree-ring and lake-sediment records of past fire
activity have strengths and weaknesses (see discussion
in Whitlock and Larsen, 2002). Tree-ring records offer
a high level of spatial resolution in a fire reconstruction, and it is possible to use tree-ring records to
distinguish between locations that experienced highseverity and low-severity fires. The location of firescarred trees identifies the exact location of particular
fires, and thus dendrochronological reconstructions
provide spatially-specific information. They are also
temporally precise in that the tree rings can be used to
date a fire event to a particular year or even a particular
season. Stand-age analysis is spatially-specific as well
(i.e., the location of the fire can be identified), but it is
temporally less precise because of the potential lag
between the fire event and the first establishment of
trees. Both methods also suffer from an ‘‘erasure
effect’’ (Agee, 1993), which means that as old trees
become progressively scarcer on the landscape the
fire-history data they contain is essentially ‘‘erased’’.
This attenuation reduces the likelihood of identifying
old fire events, and consequently most tree-ring studies show that fire events become less frequent as one
goes back in time. Although tree-ring methods extend
back to the age of the oldest living tree, this time span
is not long enough to capture major changes in
vegetation and climate, and the record is biased toward
low-severity fire events that scar trees but do not kill
them.
Charcoal records from lake sediments reconstruct
past fires with less temporal and spatial precision than
tree-ring records. Modern taphonomic studies have
shown that charcoal accumulation into a lake continues for a few years after a fire because of transportation and redeposition of secondary charcoal
within the watershed and the lake (Millspaugh and
Whitlock, 1995). This process tends to blur the exact
age of a fire, even when charcoal particles are directly
dated by radiocarbon dating. Additionally, the charcoal deposited in a lake may represent more than one
fire within the watershed, or fires from more than one
year. As a result, lake-sediment researchers refer to a
fire episode as one or more fires occurring in the time
interval of interest, rather than individual fires. Ideally,
lakes with annually laminated (i.e., varved) sediment
permit the age of a fire to be established to the year
7
(Clark, 1990), but such lakes are rare in the western US
and the timing of charcoal deposition still post-dates
the fire it represents. In non-laminated sediment
records, the age of a fire event is established by
interpolation from a series of radiometric ages determined by 210 Pb and radiocarbon dating. The radiocarbon dates are converted to calendar years, and the
possibility of multiple calendar ages for a single
radiocarbon date introduces another source of error
in the chronology. This lack of temporal precision,
however, does not prevent the identification of individual fires in areas with long fire intervals. For
example, fires in the coastal rainforests of the Pacific
Northwest are infrequent enough to be distinguished
in lakes with sedimentation rates of decades per
centimeter; in contrast, dry-forest sites with frequent
fires tend to produce less-discrete charcoal peaks.
Charcoal peaks in the uppermost sediments (dated
with 210 Pb) can be calibrated by matching them
against known fires. This modern calibration helps
to establish a value that distinguishes background
levels of charcoal from significantly higher amounts
of charcoal, i.e., charcoal peaks. Background charcoal
is the secondary material that is introduced into lakes
during non-fire years and is present in measurable
amounts in almost every centimeter of cores from the
western US. Variations in background charcoal are
thought to reflect changing levels of fuel biomass (see
Long et al., 1998; Whitlock and Larsen, 2002). In
general, a region dominated by closed mixed-conifer
forest with abundant woody fuel will produce higher
background levels of charcoal than an open parkland
with less woody fuels, although more research is
needed to define the specific relationship between
background charcoal and fuel biomass. The peaks
above the background level represent the fire events
and some component of natural variability.
Although not as precise as tree-ring records, charcoal records from lake sediments can provide some
information on the spatial location and size of the fires
represented by charcoal data. Records of macroscopic
charcoal particles (>100 mm) are interpreted as resulting from a fire within the lake’s watershed, based on
empirical evidence that particles of this size do not
travel far from their source (Whitlock and Millspaugh,
1996; Clark et al., 1998; Gardner and Whitlock, 2000).
In addition, local fires upwind of the lake are better
recorded than those downwind (Whitlock and Mill-
8
C. Whitlock et al. / Forest Ecology and Management 178 (2003) 5–21
spaugh, 1996; Gardner and Whitlock, 2000). Moreover, the size of a charcoal peak is not necessarily
indicative of the size of the fire, because fires close to
the lakeshore can deposit abundant charcoal into a
lake even if a fire is relatively small. Identification of
the potential fire source area rests on information
about recent historical fires, either from documentary
evidence or tree-ring studies.
2.2. Vegetation history reconstructions
In addition to charcoal, lake sediments also provide
long-term records of vegetation change. Pollen in lake
sediments is a primary tool for reconstructing past
variations in vegetation and climate in the western US.
The relationship between modern lake-sediment pollen and present-day vegetation and climate is used to
reconstruct past vegetation–climate relationships, taking into consideration that pollen records are affected
by the ability of some species to produce more pollen
than others and that pollen of some taxa may travel
very long distances (e.g., Fall, 1992; Davis, 1995;
Minckley and Whitlock, 2000). Pollen is generally
identified to the level of genus, but sometimes only
family-level identification is possible. Pollen identifications are supplemented by the presence of seeds,
leaves, and other plant macrofossils in the sediment,
which can often be identified to species and provide
information about local vegetation (Birks, 2002).
These macrofossils can also be radiocarbon dated to
provide a chronology for the sediment record.
2.3. Fire-related erosion
Past episodes of erosion into a lake can be measured
by subtle increases in the sediment magnetic susceptibility, which is a proxy for the introduction of ironbearing minerals from weathered bedrock, especially
magnetite, or from soils that have burned and been
thermally altered (Millspaugh and Whitlock, 1995;
Gedye et al., 2000). Non-destructive measurements of
the magnetic susceptibility of sediments help to identify time periods when clastic material is mixed with
autochthonous lake sediments. When pulses of clastic
material are associated with charcoal peaks, they can
provide evidence of local fires that were extensive and
severe enough to trigger mass-wasting events within
the watershed.
3. Modeling past and future fire conditions
3.1. Models
Our understanding of fire regimes gained from
paleoecological data can be enhanced using a variety
of climate, vegetation, and other physically-based
process models. These numerical models can simulate
the physical conditions supporting fire regimes and
thus allow us to test hypotheses about the processes
driving fire regimes at different times in the past.
Models can also be used to simulate the potential
future impacts of changing climate, vegetation, and
fire on the landscape.
Coupled atmosphere–ocean general circulation
models (AOGCMs) are used to identify the patterns
and causes of climate change (COHMAP Members,
1988; Kutzbach et al., 1998), and they have been used
to study past, present, and future climates (IPCC,
2001). AOGCMs typically have a coarse spatial resolution (a grid cell may span a few degrees of latitude
and longitude) and a relatively crude representation of
topography. They also often oversimplify the simulation of those climate variables that are most critical for
understanding plant distributions and potential fire
conditions, such as soil moisture. One of their
strengths, however, is in demonstrating the potential
effects of variations in the seasonal cycle of insolation,
ice cover, atmospheric CO2, and topography on regional climate. Output from AOGCMs can be used as
boundary conditions for regional, or mesoscale, climate models that offer a finer spatial resolution and
more elaborate depictions of the climate processes that
operate at the land surface (Giorgi and Mearns, 1991).
A variety of vegetation models have been used to
simulate past and potential future changes in vegetation.
These models range from statistical models (e.g.,
Iverson and Prasad, 1998; Shafer et al., 2001) to more
mechanistic, process-based models (e.g., Haxeltine and
Prentice, 1996). Most recent has been the development
of dynamic global vegetation models (DGVMs), such as
LPJ (Sitch et al., 2003) and IBIS (Foley et al., 1996),
which simulate complex biogeochemical processes
and vegetation dynamics including disturbance
regimes. In addition to climate and vegetation models,
other physically-based process models, such as waterbalance models, can be used to simulate important
fire-related properties, e.g., soil moisture.
C. Whitlock et al. / Forest Ecology and Management 178 (2003) 5–21
3.2. Data-model comparisons as a research
approach
When several paleoecological records are compared, it is possible to discern local, regional, and
continental-scale patterns of fire and vegetation
change. However, different physical processes may
create similar patterns of change, and the same processes may interact differently at different temporal
and spatial scales. Models can be used to test which
processes are the most plausible controls of particular
patterns, and data-model comparison efforts are
increasingly common (e.g., COHMAP (COHMAP
Members, 1988) and BIOME6000 (Prentice and
Webb, 1998)). For example, climate reconstructions
based on the paleoecological record can be compared
with paleoclimate simulations produced by AOGCMs
(Barnosky et al., 1987; Thompson et al., 1993; Bartlein et al., 1998). Such comparisons of data and model
results highlight regions where the paleoclimate
reconstructions seem well explained by a particular
set of climate variables, and also identify regions
where additional data and model development might
be useful. Paleoecological databases also provide
information on changing distributions and associations of plant and animal taxa through time (Webb,
1988; Delcourt et al., 1983; FAUNMAP Working
Group, 1996). Paleoenvironmental databases for a
variety of proxy data are now available through the
NOAA World Data Center for Paleoclimatology
(http://www.ngdc.noaa.gov/paleo/paleo.html).
4. Controls of long-term variations in fire
history in the northwestern US
4.1. Large-scale controls of climate change
In the northwestern US, the long-term controls of
climate change and how they influence regional environmental changes are relatively well known for the
last 20,000 years—the time since the last ice age. By
convention, the late-glacial period refers to the period
from 16,000 to 11,000 cal year BP (i.e., calendar years
before present) and the Holocene represents the last
11,000 cal years. Based on the timing of major
environmental changes, we divide the Holocene into
three periods for convenience of discussion: the early
9
Holocene (ca. 11,000–7000 cal year BP), middle
Holocene (7000–4000 cal year BP), and late Holocene
(4000 cal year BP to present). The actual timing of
Holocene environmental changes varies somewhat
from region to region.
In the last 20,000 years, the large-scale controls of
climate included the size and position of the Laurentide and Cordilleran ice sheets and their influence on
temperature gradients and the strength and position of
the jet stream (and thus winter storm tracks; Thompson et al., 1993; Bartlein et al., 1998). The size of the
ice sheets also determined the strength of the glacial
anticyclone and the flow of surface winds off the ice
sheet. Another important control of regional climate
has been the variations in the amplitude of the seasonal
cycle of insolation, which have occurred as a result of
regular variations in the earth–sun orbital geometry,
the tilt of the earth on its axis, and the timing of
perihelion (the point in the earth’s orbit when it is
closest to the sun). These orbital variations occur on
scales of 104 to 105 years, and they are considered the
pacemaker of glacial–interglacial cycles (Imbrie et al.,
1984). During the last glacial maximum approximately 20,000 cal year BP, summer and winter insolation were similar to their present values, but at 11,000
cal year BP summer insolation was 8% higher and
winter insolation was 8% lower than at present at 458N
latitude. This increased seasonality led to increased
summer temperatures and decreased effective moisture (i.e., precipitation minus evapotranspiration), and
indirectly influenced the strength of the northeast
Pacific subtropical high-pressure system and summer
monsoons (Thompson et al., 1993; Whitlock and
Bartlein, 1993). The fact that many types of paleoenvironmental records from the western US register a
response to variations in insolation on orbital time
scales attests to the importance of insolation variations
in shaping regional climate and vegetation conditions
during the late-glacial and Holocene periods.
The post-glacial vegetation history of the western
US is described in several recent reviews (Baker,
1983; Adam, 1985; Mehringer, 1985; Barnosky
et al., 1987; Anderson, 1990; Whitlock, 1992, 1993;
Thompson et al., 1993; Fall et al., 1995). The lateglacial period was a time of biotic reorganization in
the northwestern US as the influence of large ice
sheets on climate was waning and summer insolation
was increasing (Whitlock, 1992). Plant taxa colonized
10
C. Whitlock et al. / Forest Ecology and Management 178 (2003) 5–21
regions vacated by glaciers, creating new pioneer
communities. Regions outside the glacial limits were
also undergoing rapid change, and in many areas the
vegetation combined present-day subalpine and lowland taxa in assemblages that have no modern counterpart (e.g., Whitlock, 1992, 1993). As climate
continued to warm, species distributions became
arranged along gradients of latitude and elevation.
The seasonal cycle of insolation was greatest in the
early Holocene and led to warmer summers and
probably colder winters than at present throughout
the region. The northeastern Pacific subtropical highpressure system was stronger, and consequently the
Pacific Northwest and much of the northern Rocky
Mountains that are summer-dry (i.e., that receive most
of their annual precipitation in winter) were even drier
in summer in the early Holocene (Whitlock and
Bartlein, 1993). Greater summer drought in the early
Holocene is evidenced by the expansion of xerophytic
communities, higher alpine treeline, and lower lake
levels and stream activity. The summer monsoon of
the southwestern US and parts of the Rocky Mountains also intensified in the early Holocene, and areas
that today are summer-wet (i.e., that receive most of
their annual precipitation in summer) were even wetter
then, as indicated by the expansion of mesophytic
plant taxa in those regions (Thompson et al., 1993;
Whitlock and Bartlein, 1993).
Summer insolation decreased and winter insolation
increased to present-day levels in the last 7000 years,
and this attenuation of the seasonal cycle of insolation
eventually led to the establishment of the modern
vegetation, climate, and fire regimes. Cooler and
effectively wetter conditions than before were established in the Pacific Northwest and Rocky Mountains
between 5000 and 3000 cal year BP, as evidenced by
increased mesophytic vegetation and downslope shifts
in alpine treeline. Summer-wet regions record a
decrease in summer precipitation in the middle and
late Holocene as a result of reduced summer onshore
flow of moisture. This drying led to the establishment
of more-open xerophytic communities than before
(Thompson et al., 1993).
Superimposed on climate changes driven by orbital
variations are short-term variations in climate, such as
the Medieval Warm Period (ca. 900–600 cal year BP;
Gates, 1993) and the Little Ice Age (ca. 550–150 cal
year BP; Gates, 1993). The cause of these variations is
not known, but changes in solar output and volcanic
activity have been suggested (e.g., Crowley, 2000;
Bond et al., 2001). Tree-ring and lake-sediment
records in the Sierra Nevada (Stine, 1994; Hughes
and Graumlich, 1996), for example, show evidence of
drought during the Medieval Warm Period, and on
many mountain ranges treeline lay above present
levels (Leavitt, 1994). Cooler temperatures and
expanded glacial activity are associated with the Little
Ice Age in the northern Rocky Mountains and the
Pacific Northwest (Davis, 1988), and this period
seems to be the coldest of a series of cool oscillations
in the late Holocene (e.g., Luckman, 1994).
4.2. Long-term fire responses
A network of lake-sediment records provides a
picture of changing fire activity in different parts of
the northwestern US through the Holocene (Fig. 1).
Various approaches have been used to reconstruct fire
frequency (e.g., see Mehringer et al., 1977; Cwynar,
1987; Mehringer, 1985; Fall, 1997), but we explicitly
consider sites where macroscopic charcoal was examined in contiguous samples. As noted above, such
studies constrain the source of the fire and are calibrated based on a comparison with dendrochronological and historical records. Also, by distinguishing
charcoal peaks from the slowly varying background
trend, it is possible to examine changes in the frequency of fire episodes through time.
The records suggest that fire frequency (i.e., the
return interval between fires) has changed continuously and no long-term stationarity is evident in the
fire frequency record (Fig. 2). There are also distinct
features in the spacing of fire episodes within regions
that suggest spatial coherency in the disturbance
regimes. Charcoal records from the summer-dry Klamath Mountains of northern California, for example,
indicate that fires were frequent throughout the Holocene with little long-term trend in the data. Two sites
from the Klamath region, Bluff and Crater lakes,
suggest periods of frequent fire, such as at ca. 4000
and 1000 cal year BP. These periods may have been
times of pronounced drought in northern California
(Mohr et al., 2000). In contrast to the fire records, the
pollen records from these sites record a large response
to insolation-driven climate changes, and little change
in association with the short-term variations in climate.
C. Whitlock et al. / Forest Ecology and Management 178 (2003) 5–21
11
Fig. 1. Location of fire-history studies referred to in Fig. 2: Cedar Lake, CA (Whitlock et al., unpublished data); Bluff Lake and Crater Lake,
CA (Mohr et al., 2000); Taylor Lake, OR (Long and Whitlock, 2002); Little Lake, OR (Long et al., 1998); Cygnet Lake, WY (Millspaugh
et al., 2000); Trail Lake, WY (Whitlock, unpublished data); Slough Creek Lake, WY (Millspaugh and Whitlock, in press); Pintlar Lake, MT
(Brunelle-Daines, 2002).
For example, an expansion of xerophytic taxa, most
notably Quercus spp., is consistent with warmer drier
conditions in the early Holocene, but no change in fire
frequency is registered on this time scale (Mohr et al.,
2000). In contrast, pollen data were not sensitive
enough to detect vegetation changes on millennial
and shorter time scales in this area, when the region
has experienced elevated fire activity.
Data from the summer-dry Oregon Coast Range
show high fire activity in the early Holocene and a
trend towards fewer fires in the late Holocene (Long
et al., 1998; Long and Whitlock, 2002). Fire-sensitive
species, such as Douglas-fir, red alder (Alnus rubra),
Oregon white oak (Quercus garryana), and bracken
(Pteridium spp.), are well represented in the early
Holocene interval. The decrease in fire activity correlates well with an expansion of mesophytic rainforest
taxa, such as Sitka spruce (Picea sitchensis) and
western hemlock (Tsuga heterophylla), and provides
supporting evidence of cooling and increased moisture
in the last few millennia.
Charcoal records from the summer-dry region of
southern Yellowstone National Park also indicate
increased fire activity associated with warmer and
drier conditions in the early Holocene (Millspaugh
et al., 2000). Two sites, Cygnet and Trail lakes,
register this fire response, and it occurs even where
the vegetation has been relatively unresponsive to
Holocene climate changes. At Cygnet Lake, for example, infertile, well-drained soils have maintained lodgepole pine forests over the last 11,000 years
(Whitlock, 1993). In contrast, Trail Lake on andesitic
terrain records an expansion of Douglas-fir and lodgepole pine in the early Holocene followed by increased
spruce, fir, and whitebark pine in the late Holocene. At
Cygnet Lake, the fire frequency was low prior to
11,000 cal year BP during an initial tundra-open
meadow period. With the arrival of lodgepole pine,
12
C. Whitlock et al. / Forest Ecology and Management 178 (2003) 5–21
Fig. 2. Comparison of Holocene fire reconstructions from sites in different geographic and climatic locations in the western US (see Fig. 1 for
site locations). The horizontal lines at each site represent past fire episodes, i.e., based on the age of peaks in the charcoal record obtained from
radiocarbon-dated sediment cores (see Whitlock and Larsen, 2002). These peaks represent one or more fires occurring in a decade. The curves
depict fire frequency, which is the number of fire episodes per 1000 years averaged over a moving window. Summer dry and summer wet refer
to two different precipitation regimes evident in the western US (see text for discussion).
fire frequency increased and reached a maximum of
10–15 fires per 1000 years at 10,000–9000 cal year BP.
In the last 7000 cal year BP the fire frequency declined
to a modern return interval of 2–3 fires per 1000 years
in the last two millennia, but lodgepole pine remained
the dominant forest species at this site. This record
shows that fire variations respond directly to climate
even in areas where the vegetation is fairly unresponsive. Trail Lake indicates a decline in fire frequency
from the early Holocene period, with an increase ca.
1000 cal years ago.
Records from summer-wet areas of the Rocky
Mountains provide additional evidence of the strong
link between climate and fire conditions. These locations are areas that were influenced by increased
summer precipitation in the early Holocene and have
become progressively drier since then. Slough Creek
Lake in northwestern Wyoming and Pintlar Lake in
southwestern Montana register this response in their
pollen and charcoal records (Brunelle-Daines, 2002;
Millspaugh and Whitlock, in press). Both sites have
fire frequencies that were low in the early Holocene
and became progressively more frequent in the late
Holocene. At Slough Creek Lake, this increase in fire
activity is associated with the expansion of Douglas-fir
parkland at low elevations, replacing pine forest.
C. Whitlock et al. / Forest Ecology and Management 178 (2003) 5–21
At Pintlar Lake, lodgepole pine forest replaced mixedconifer forest.
Millspaugh and Whitlock (in press) compare the
Slough Creek Lake and Cygnet Lake records on
shorter time intervals and show that, while the fire
histories in summer-wet and summer-dry regions were
out of phase on orbital time scales, they were occasionally synchronized on short time scales. For example, fire episodes at these sites are registered at ca.
1000, 2000, and 4000 cal year BP, and these particular
events are associated with fire-related debris flow
activity in northeastern Yellowstone National Park
(Meyer et al., 1995). This synchronization of fires
in the late Holocene may have been enhanced by the
effect of decreasing summer insolation on atmospheric circulation and precipitation regimes. The
climate of summer-wet and summer-dry regions
became more similar in the late Holocene, as summer-dry areas became wetter and summer-wet areas
became drier. During this period, the large-scale controls that cause different fire regimes on orbital time
scales were overridden by widespread droughts and
fires on shorter time scales. For example, fires have
occurred in both summer-wet and summer-dry areas in
recent years in the face of dry conditions.
5. Potential future fire conditions in the
northwestern US
5.1. Fire regimes and climate change
The relationships among climate, vegetation, and
fire regimes expressed in the paleoecological record,
combined with model simulations of past, present, and
potential future environmental changes, can help us
understand how fire regimes may change in the future
in the northwestern US. Projecting potential future fire
regimes requires accurately simulating a myriad of
complex ecological, physical, and human interactions,
a task that is not yet possible. We can, however,
examine the magnitude and direction of simulated
changes in the physical and biotic environment and
infer their potential impacts on fire regimes. This
information, combined with an appreciation of the
historical range of variability in fire occurrence, can
help develop appropriate management strategies for
the region.
13
Although there is still uncertainty in model simulations of future climate conditions, it is increasingly
evident that human activities are contributing to climate changes that will significantly affect future fire
regimes (IPCC, 2001). A recent study by Giorgi et al.
(2001) compared the regional results of transient
climate simulations from nine AOGCMs for the time
period A.D. 2071–2100. Each AOGCM was run using
two different greenhouse gas emission scenarios, with
each scenario assuming different rates of increase in
greenhouse gases (see Giorgi et al., 2001; IPCC, 2001
for descriptions of the emission scenarios). All of the
AOGCMs simulated global warming with mean
annual temperature increases ranging from approximately 0.9 to 4.5 8C among the models (Giorgi et al.,
2001). Of particular significance is the general agreement of the AOGCM results—at least seven of the
AOGCMs under each of the emission scenarios simulated warming in summer and winter for the northwestern US that was greater than the simulated global
average annual warming. As Giorgi et al. (2001) note,
the agreement of the model simulations under two
different emission scenarios increases the confidence
that can be ascribed to the simulations of regional
warming. Although potential changes in precipitation
are more difficult to simulate than those in temperature, at least seven of the AOGCM simulations for A.D.
2071–2100 also indicated slight increases in winter
precipitation for the northwestern US. There was less
agreement among the AOGCMs as to whether summer
precipitation would increase or decrease in the region
(Giorgi et al., 2001).
There are many ways in which these potential
future climate changes could affect fire regimes.
Significant increases in summer temperatures in the
northwestern US without an accompanying increase in
precipitation could increase summer drought stress on
vegetation. If the drought stress was severe enough it
could create vegetation conditions conducive to
increased fire activity. Even if summer precipitation
in the region increased significantly, greater evapotranspiration as a result of higher temperatures could
still lead to increased drought stress. If drought stresses were large enough, they could affect riparian
vegetation, which is often buffered from drought
effects, particularly along perennial streams. Changes
in temperature and the timing of precipitation would
also affect fuel moisture levels, with drier fuels
14
C. Whitlock et al. / Forest Ecology and Management 178 (2003) 5–21
increasing the potential for fires (Miller and Urban,
1999).
Increased temperatures could also result in an earlier and longer fire season (Flannigan et al., 2000;
Gitay et al., 2001). Flannigan et al. (2000) evaluated
simulations by two AOGCMs for 2060 using scenarios
that assumed an approximate doubling of atmospheric
CO2 concentrations. They concluded that both
AOGCMs simulated an increase in fire activity for
the northwestern US based on an increase in the
seasonal fire-severity rating index, which they defined
as ‘‘a seasonal mean of the daily estimate of the
control difficulty of a potential fire’’ (Flannigan
et al., 2000, p. 222). Potential future changes in
climate could also affect the frequency of natural
ignition sources such as lightning. Although it is
difficult to extrapolate from long-term climate
changes simulated by AOGCMs to relatively shortterm weather phenomena, such as thunderstorms,
simulations of future climate change indicate a potential increase in atmospheric conditions that promote
lightning-caused fires (e.g., Price and Rind, 1994).
As one measure of the potential impacts of future
climate change on fire regimes in the northwestern US,
we used a simple water-balance model to simulate
potential changes in summer soil moisture for A.D.
2050–2059. We also simulated summer soil moisture
for 6000 cal year BP, a time period when many
paleoecological records from the northwestern US
register increased fire frequencies. Soil moisture is
an important control on vegetation, and it integrates
changes in temperature, precipitation, and evapotranspiration, providing a measure of potential drought
stress. The simulations for 6000 cal year BP and A.D.
2050–2059 show similar patterns of soil moisture,
particularly in the mountains of the northwestern
US (Fig. 3). For 6000 cal year BP, the paleoecological
record indicates that increased summer insolation
resulted in higher-than-present summer temperatures,
heightened drought, and drier soils and fuels. In turn,
these changes led to fire frequencies that were greater
than current levels in summer-dry regions. The simulated soil moisture anomalies for 6000 cal year BP
(Fig. 3) are in agreement with the paleoecological
record and indicate that much of the northwestern US
had drier-than-present summer soil moisture at this
time. Drier-than-present soils are also simulated for
A.D. 2050–2059 as a result of simulated increases in
summer temperatures and evapotranspiration, and, in
some areas of the northwestern US, earlier snowmelt.
Fig. 3. Mean summer (July–September) soil moisture anomalies (mm) for 6000 cal year BP (left) and A.D. 2050–2059 (10-year mean) (right)
compared to present (A.D. 1951–1980, 30-year mean) in the northwestern US. The Olympic Mountains, Cascade Range, and Rocky Mountains
display similar patterns of negative summer soil moisture anomalies for both time periods, indicating slightly drier soils in summer in these
regions as compared with present conditions. Climate data: A.D. 1951–1980 (Thompson et al., 1999); 6000 cal year BP, CCM3 (Bonan, 1996);
A.D. 2050–2059, HADCM2 HCGSa (Mitchell and Johns, 1997). Soil data: CONUS-Soil (Miller and White, 1998).
C. Whitlock et al. / Forest Ecology and Management 178 (2003) 5–21
There are also large areas of the interior northwestern
US for which small increases in soil moisture are
simulated as a result of simulated increases in summer
precipitation.
The similar patterns of simulated summer soil
moisture anomalies for 6000 cal year BP and A.D.
2050–2059 suggest that future climate changes could
create soil moisture conditions similar to those that
occurred during times in the Holocene when fire
frequencies were higher. That the simulated patterns
of soil moisture for the two time periods are similar
does not mean that the response of the fire regime
would be same in the future as at 6000 cal year BP,
since other factors, including vegetation, affect fire
regimes. However, by continuing to compare paleoecological data with model simulations of soil moisture
and other variables, we can improve our understanding
of both how and why interactions among climate,
vegetation, and fire regimes have varied in the past,
which in turn provides important information as to
how these interactions may vary in the future.
5.2. Fire regimes and vegetation change
Future vegetation changes in the northwestern US
will also affect fire regimes. Vegetation change can
occur as a result of changes in climate, interactions
among native and introduced species, and human
activities, such as land use. Future changes in temperature and in the magnitude and frequency of precipitation events will have significant impacts on
biotic communities in the northwestern US. Changes
in seasonality will be particularly important. For
example, warmer temperatures, if combined with
increases in precipitation, could increase plant productivity in certain regions, providing more fuel during the fire season (Gitay et al., 2001). In addition to
responding to future changes in temperature and precipitation, plants are also directly affected by increases
in atmospheric concentrations of CO2, which increase
the drought tolerance of some plant species by
improving their water-use efficiency (Mooney et al.,
1999; Simberloff, 2000). Vegetation models that
incorporate anticipated increases in atmospheric
CO2 concentrations simulate an expansion of woody
vegetation (e.g., Shafer, 2000), particularly in arid
regions, that could potentially increase fuel biomass
and fuel connectivity in these areas.
15
The paleoecological record indicates that plant taxa
respond individualistically to climate change (Davis,
1981; Huntley, 1995; Webb, 1995) and this response is
also observed in the modern record (e.g., Fitter and
Fitter, 2002). As climate changes, the ranges of some
plant species will expand and others will contract,
altering the species composition of communities (Shafer et al., 2001). The potentially disparate response of
species to changes in climate and atmospheric CO2
concentrations will alter competitive interactions and
affect the geographic distribution of fire-tolerant and
fire-intolerant species. In turn, the changing mosaic of
species on the landscape will alter the frequency and
extent of future fires. In model simulations, Miller and
Urban (1999) found that species composition also
significantly affected fuel-bed bulk density, which
in turn affects fire regimes.
Interactions of plant species with non-plant taxa will
also affect fire regimes. For example, increases in
potential winter temperatures will allow some insect
species to significantly expand their ranges, while
increased summer temperatures will increase the productivity of some species (Ayres and Lombardero,
2000). Potential increases in the frequency and magnitude of droughts could increase stress on trees making
them more vulnerable to insect attack, and damage from
insect outbreaks would leave stands more vulnerable to
fire (Knight, 1987). The fire-insect outbreak feedback is
one of the many climate–vegetation–fire interactions
that will likely be more evident in the future.
Human activities, including the introduction of nonnative invasive species, and land use changes, such as
logging and habitat fragmentation, will affect the
vulnerability of forest communities to climate change.
Human introductions of certain non-native invasive
species, such as cheatgrass (Bromus tectorum), have
already altered fire regimes in portions of the northwestern US, and these plants are continuing to spread
(Mack, 1981). The invasion of cheatgrass in intermountain regions of the western US has created a
continuous cover of fine fuels in some areas that were
formerly covered by patchy bunchgrasses, allowing
fires to spread more easily (West and Young, 2000).
Fires also create opportunities for invasive species to
establish (D’Antonio, 2000).
The paleoecological record and future vegetation
simulations suggest that certain plant species are relatively resistant to climate changes, while others respond
16
C. Whitlock et al. / Forest Ecology and Management 178 (2003) 5–21
to climate change with major geographic readjustments. Lodgepole pine, for example, may be relatively
resistant to projected changes in climate and fire frequency just as it has shown little response to warming in
the past (Whitlock, 1993; Bartlein et al., 1997). With
increasing fire activity in the future, the structure of
lodgepole pine forests may shift to younger age classes
and a finer-grained mosaic (Romme and Turner, 1991),
but the forest dominant could remain the same. In
contrast, Douglas-fir, which did well during periods
of warmer-than-present summer conditions and higher
fire frequencies in the past, may be especially vulnerable to projected climate changes that include warm
winters. Warmer winter temperatures could reduce the
competitiveness of its seedlings by not providing a
sufficient period of winter cold temperatures to meet
the chilling requirement of this species (McCreary et al.,
1990; Kirschbaum et al., 1996). Moreover, Douglas-fir
persisted during times of increased fire occurrence, but
model projections suggest that it will decline if frequencies exceed 1 fire per 20 years (Keane et al., 1990).
The simulated distribution of Douglas-fir for A.D.
2090–2099 (Fig. 4) illustrates the potential spatial
complexity of the vegetation response. The response
surface model used in this simulation identifies areas in
the northwestern US where potential future bioclimatic
conditions are similar to those currently occupied by
Douglas-fir, as well as areas where future bioclimatic
conditions are sufficiently different from current conditions that they may no longer be suitable for the
species. The model results indicate a potential reduction
in suitable bioclimatic conditions for Douglas-fir along
the Pacific slope and western margins of its range, with
potential new areas of suitable bioclimatic conditions
along the eastern margins of its range. There are a
number of caveats that accompany this model simulation, including the recognition that there are many
processes that will affect the response of Douglas-fir
to future climate change that are not simulated by the
response surface model. These processes include the
response of the species to changing atmospheric CO2
concentrations, interactions with other species, and the
Fig. 4. Potential future distribution of P. menziesii (Douglas-fir) in the northwestern US, simulated using climate scenarios for A.D. 2090–2099
(10-year mean) generated by the HADCM2 AOGCM (Mitchell and Johns, 1997). ‘‘No change’’ indicates where P. menziesii is observed at
present and is simulated to occur under future climate conditions; ‘‘contraction’’ indicates where P. menziesii is observed at present but is
simulated to be absent under future climate conditions; and ‘‘extension’’ indicates where P. menziesii is not observed at present but is
simulated to occur under future climate conditions. Climate data: A.D. 2090–2099 HADCM2 HCGSa (Mitchell and Johns, 1997). See Shafer
et al. (2001) for methods and additional species. Modified from Shafer et al. (2001).
C. Whitlock et al. / Forest Ecology and Management 178 (2003) 5–21
effects of changes in disturbance regimes, such as fire
(Shafer et al., 2001). Thus, the response surface model
results are not predictions of range changes at any
particular location on the ground. Rather, the results
suggest that bioclimatic conditions over large areas of
its current range may become unsuitable as a result of
projected climate changes. Future fires in these areas
could initiate the conversion of Douglas-fir forests to
those dominated by different species if future bioclimatic conditions limit the establishment and viability of
Douglas-fir seedlings.
Finally, a critical point to remember in assessing the
potential future impacts of climate change on fire
regimes and vegetation is that climate change is transient—as global warming continues to increase the
interactions among climate, vegetation and fire regimes
will also change. Increased winter precipitation, for
example, is simulated to increase snowpack in the
Yellowstone region of the northwestern US for A.D.
2050–2059. As winter temperatures continue to warm,
however, a point will be reached where less precipitation
will occur as snow, and snowpack will begin to decrease.
These changes will affect the timing and quantity of
snowmelt, which in turn will affect soil moisture,
streamflow, and upland and riparianvegetation in certain
areas. The transient nature of climate change greatly
increases the difficulty for managers of developing
appropriate responses to future climate change.
6. Relevance to ecosystem management
An understanding of climate, vegetation, and fire
interactions in the past provides important perspectives for management efforts promoting future sustainability of forests and riparian ecosystems. First, the
paleoecological record from the northwestern US
shows that fire frequency has continually changed
during the Holocene in response to climate change.
Periods of intense drought in the past featured more
fires and fire-adapted vegetation, and wet conditions
were associated with fewer fire events and the establishment of more fire-sensitive forests. The timing of
drought and high fire activity has varied from region to
region, reflecting the spatial heterogeneity of the
seasonal patterns of precipitation at any particular
time (Mock, 1996). Of importance for ecosystem
management is the lack of a fire cycle or recurrent
17
fire interval at centennial and millennial time scales
based on charcoal data. Likewise, tree-ring studies in
the Pacific Northwest and northern Rocky Mountains
show considerable temporal variability in fire occurrence on decadal time scales (Heyerdahl et al., 1998;
Weisberg and Swanson, 2003).
The variability recorded in the fire-history record is
important because it contradicts the notion of a static
fire return interval that is often referred to in ecosystem management plans. Although mean fire interval
may be a useful concept for management planning in
certain instances, such as in defining an envelope of
natural variability (Swetnam et al., 1999), this concept
should be used cautiously. The fire activity of the last
few centuries in many areas of the western US is not
typical of the longer record. Overcoming the effects of
fire suppression in recent decades is a worthy goal, but
it is important to remember that even our best efforts
cannot reproduce pre-Euro-American settlement fire
regimes. The climate, vegetation, and fire conditions
that existed then do not exist today and are not likely to
exist in the near future. Although the past does not
provide a perfect analog for the future, knowledge of
previous fire regimes, combined with our understanding of modern fire–climate relations, can help us
anticipate future changes in fire regimes. For example,
projections of increased summer drought suggest that
fire frequency or fire severity could increase in most
western US forests (see also Meyer and Pierce, 2003).
Second, northwestern US forests, including riparian
forest communities, are relatively young. Present
communities have formed only in the last few millennia as the region’s climate became cooler and wetter
(Whitlock, 1992). Old-growth riparian communities
may have established during a cooler wetter period of
generally low fire activity during the Little Ice Age or
following particularly widespread fires in the late
1500s (Weisberg and Swanson, 2003). Thus, the forests that greeted early Euro-American settlers to the
region would be difficult to recreate under present
climate and fire conditions.
Third, the individualistic behavior of the region’s
dominant tree species has led to range contractions
and expansions as well as entire shifts in species
distributions. Given the magnitude of projected climate changes, individualistic responses of species
may be more pronounced in the future. The paleoecological record reveals that shifts in fire regimes have
18
C. Whitlock et al. / Forest Ecology and Management 178 (2003) 5–21
sometimes been the catalyst of rapid vegetation
changes in the past, and it is likely that disturbance
will influence how vegetation adjust to future climate
changes. Certain species have persisted in the past
under higher fire frequencies than they experience
today, and these species may be able to compete
favorably if fire frequencies increase in the future.
Fourth, different sources of fire-history data help us
to evaluate fire–vegetation–climate relationships on
multiple time scales and across landscapes. Fire reconstructions based on lake-sediment records, dendrochronological data, and documentary evidence help
disclose the temporal and spatial variability of fire in
ecosystems. For example, in low-elevation forests that
experienced frequent fires in recent times, the effects
of fire suppression and other management practices
are well recorded in tree-ring data (e.g., Hessburg and
Agee, 2003). Charcoal records from such forests
(although currently not available) would disclose
the frequency of stand-replacing events that are lethal
to the trees, information that the tree-ring record is
unable to provide. In montane and subalpine forests,
current fire regimes include mixed- to high-severity
infrequent fires. Dendrochronological data poorly
reconstruct the frequency of such events, whereas
the charcoal data provide information on stand-replacing fires spanning several millennia.
Fifth, an understanding of the spatial heterogeneity
of precipitation regimes in the northwestern US in the
past offers another tool for forward-looking ecosystem
management. Future fire conditions in summer-dry
regions may be best described by an examination of
the early Holocene, when drier-than-present summers
resemble those of future projections. In summer-wet
sites, the last few millennia are the time of highest fire
occurrence and may be the best analog for the future
(although this assertion is based on very few records and
needs further testing). Future trade-offs in the spatial
distribution of precipitation will particularly alter riparian ecosystems by affecting the timing and magnitude
of discharge, including snowmelt and storm events; the
frequency and severity of fire on riparian communities;
and the potential for invasion by non-native species
(Dunham et al., 2003). Our ability to describe the spatial
extent and character of future conditions will continue
to improve as climate and vegetation models increase in
resolution and complexity, and disturbance regimes are
more explicitly addressed.
Finally, an important challenge is to incorporate the
complexity of climate–vegetation–fire relationships
into a conceptual framework of fire management. In
protecting areas of special concern, such as riparian
buffers, we must recognize the dynamic nature of
ecosystems in the past and the complexity of potential
future responses. Fire management activities that alter
the fire regime directly or indirectly may collide with
future vegetation and climate changes that are not
under management control. These interactions have
the potential to produce unexpected and undesirable
consequences.
Thus, ecosystem planning and management must be
adaptive enough to incorporate flexibility in the face of
environmental change. Adaptive management requires
clearly defined goals predicated on information gained
from research that is scientifically sound, comprehensive, and multidisciplinary (National Research Council, 2002). Long-term monitoring is necessary to
determine the successes and consequences of management decisions in the light of the complex ecological
responses that are likely in the future. The different
configurations of climate, hydrology, vegetation, and
fire in the paleoecological record provide an important
tool for delineating the magnitude of possible future
responses of fire regimes to climate and vegetation
changes, and help to identify those variables that should
be part of future ecosystem monitoring efforts.
Acknowledgements
Support for this research was provided by NASA
Earth System Science Fellowship 97/0281, and US
National Science Foundation grants ATM-9910638,
EAR-9906100, and ATM-0117160. We thank Patrick
Bartlein for assistance with data and figures, and
Grant Meyer, Daniel Muhs, Laura Strickland, and
one anonymous reviewer for their comments on the
manuscript.
References
Adam, D.P., 1985. Quaternary pollen records from California. In:
Bryant Jr., V.M., Holloway, R.G. (Eds.), Pollen Records of
Late-Quaternary North American Sediments. American Association of Stratigraphic Palynologists Foundation, Austin, TX,
pp. 125–140.
C. Whitlock et al. / Forest Ecology and Management 178 (2003) 5–21
Agee, J.K., 1993. Fire Ecology of Pacific Northwest Forests. Island
Press, Washington, DC, 493 pp.
Anderson, R.S., 1990. Holocene forest development and paleoclimates within the central Sierra Nevada, California. J. Ecol. 78,
470–489.
Arno, S.F., Sneck, K.M., 1977. A method for determining fire
history in coniferous forests in the mountain West. USDA
Forest Service General Technical Report No. INT-42.
Ayres, M.P., Lombardero, M.J., 2000. Assessing the consequences
of global change for forest disturbance from herbivores and
pathogens. Sci. Total Environ. 262, 263–286.
Baker, R.G., 1983. Holocene vegetational history of the western
United States. In: Wright Jr., H.E. (Eds.), Late-Quaternary
Environments of the United States. University of Minnesota
Press, Minneapolis, MN, pp. 109–127.
Barnosky, C.W., Anderson, P.M., Bartlein, P.J., 1987. The northwestern US during deglaciation; vegetational history and
paleoclimatic implications. In: Ruddiman, W.F., Wright Jr.,
H.E. (Eds.), North America and Adjacent Oceans During the
Last Deglaciation. Geology of North America, vol. K-3.
Geological Society of America, Boulder, CO, pp. 289–321.
Bartlein, P.J., Whitlock, C., Shafer, S.L., 1997. Future climate in
the Yellowstone National Park region and its potential impact
on vegetation. Conserv. Biol. 11, 782–792.
Bartlein, P.J., Anderson, K.H., Anderson, P.M., Edwards, M.E.,
Mock, C.J., Thompson, R.S., Webb, R.S., Webb III, T.,
Whitlock, C., 1998. Paleoclimate simulations for North
American over the past 21,000 years: features of the simulation
climate and comparisons with paleoenvironmental data. Quat.
Sci. Rev. 17, 549–585.
Birks, H.H., 2002. Plant macrofossils. In: Smol, J.P., Birks, H.J.B.,
Last, W.M. (Eds.), Tracking Environmental Change Using Lake
Sediments, vol. 3: Terrestrial, Algal, and Siliceous Indicators.
Kluwer Academic Publishers, Dordrecht, The Netherlands,
pp. 49–74.
Bonan, G.B., 1996. A land surface model (LSM version 1.0) for
ecological, hydrological, and atmospheric studies: technical
description and user’s guide. NCAR Technical Note No.
NCAR/TN-417 þ STR, Boulder, CO, 150 pp.
Bond, G., Kromer, B., Beer, J., Muscheler, R., Evans, M.N.,
Showers, W., Hoffmann, S., Lotti-Bond, R., Hajdas, I., Bonani,
G., 2001. Persistent solar influence on North Atlantic climate
and deep-ocean flow south of Iceland. Nature 297, 515–517.
Brunelle-Daines, A., 2002. Holocene changes in fire, climate, and
vegetation in the northern Rocky Mountains of Idaho and
western Montana. Ph.D. Dissertation. University of Oregon,
Eugene, OR, 178 pp.
Clark, J.S., 1990. Fire and climate change during the last 750 years
in northwestern Minnesota. Ecol. Monogr. 60, 135–159.
Clark, J.S., Lynch, J., Stocks, J.B., Goldammer, J., 1998.
Relationships between charcoal particles in air and sediments
in west-central Siberia. Holocene 8, 19–29.
COHMAP Members, 1988. Climate changes of the last 18,000
years: observations and model simulations. Science 241, 1043–
1052.
Crowley, T.J., 2000. Causes of climate change over the last 1000
years. Science 289, 270–277.
19
Cwynar, L.C., 1987. Fire and the forest history of the North
Cascade Range. Ecology 68, 791–802.
D’Antonio, C., 2000. Fire, plant invasions, and global changes. In:
Mooney, H.A., Hobbs, R.J. (Eds.), Invasive Species in a
Changing World. Island Press, Washington, DC, pp. 65–93.
Davis, M.B., 1981. Quaternary history and the stability of forest
communities. In: West, D.C., Shugart, H.H., Botkin, D.B.
(Eds.), Forest Succession: Concepts and Application. Springer,
New York, pp. 132–153.
Davis, P.T., 1988. Holocene glacier fluctuations in the American
cordillera. Quat. Sci. Rev. 7, 129–158.
Davis, O.K., 1995. Climate and vegetation patterns in surface
samples from arid western USA: application to Holocene
climatic reconstructions. Palynology 19, 95–117.
Delcourt, H.R., Delcourt, P.A., Webb III, T., 1983. Dynamic plant
ecology: the spectrum of vegetational change in time and space.
Quat. Sci. Rev. 1, 153–175.
Dunham, J.B., Young, M.K., Rieman, B.E., Gresswell, R.E., 2003.
Effects of fire on fish populations: landscape perspectives on
persistence of native fishes and non-native fish invasions. For.
Ecol. Manage. 178 (1–2), 183–196.
Fall, P.L., 1992. Pollen accumulation in a montane region of
Colorado, USA: a comparison of moss polsters, atmospheric
traps, and natural basins. Rev. Palaeobot. Palynol. 72, 162–197.
Fall, P.L., 1997. Fire history and composition of the subalpine
forest of western Colorado during the Holocene. J. Biogeogr.
24, 309–325.
Fall, P.L., Davis, P.T., Zielinski, G.A., 1995. Late Quaternary
vegetation and climate of the Wind River Range, Wyoming.
Quat. Res. 43, 393–404.
FAUNMAP Working Group, 1996. Spatial response of mammals to
late Quaternary environmental fluctuations. Science 272, 1601–
1605.
Fitter, A.H., Fitter, R.S.R., 2002. Rapid changes in flowering time
in British plants. Science 296, 1689–1691.
Flannigan, M.D., Stocks, B.J., Wotton, B.M., 2000. Climate change
and forest fires. Sci. Total Environ. 262, 221–229.
Foley, J.A., Prentice, I.C., Ramankutty, N., Levis, S., Pollard, D.,
Sitch, S., Haxeltine, A., 1996. An integrated biosphere model of
land surface processes, terrestrial carbon balance, and vegetation dynamics. Glob. Biogeochem. Cy. 10, 603–628.
Gardner, J.J., Whitlock, C., 2000. Charcoal accumulation following
a recent fire in the Cascade Range, northwestern USA, and its
relevance for fire-history studies. Holocene 5, 541–549.
Gates, D.M., 1993. Climate Change and its Biological Consequences. Sinauer Associates, Massachusetts.
Gavin, D.G., 2000. Holocene fire history of a coastal temperate rain
forest, Vancouver Island, British Columbia, Canada. Ph.D.
Dissertation. University of Washington, Seattle, WA, 131 pp.
Gedye, S.J., Jones, R.T., Tinner, W., Ammann, B., Oldfield, F.,
2000. The use of mineral magnetism in the reconstruction of
fire history: a case study from Lago di Origlio, Swiss Alps.
Palaeogeogr. Palaeoclimatol. Palaeoecol. 164, 101–110.
Giorgi, F., Mearns, L.O., 1991. Approaches to the simulation of
regional climate change: a review. Rev. Geophys. 29, 191–216.
Giorgi, F., Whetton, P.H., Jones, R.G., Christensen, J.H., Mearns,
L.O., Hewitson, B., vonStorch, H., Francisco, R., Jack, C.,
20
C. Whitlock et al. / Forest Ecology and Management 178 (2003) 5–21
2001. Emerging patterns of simulated regional climatic changes
for the 21st century due to anthropogenic forcings. Geophys.
Res. Lett. 28, 3317–3320.
Gitay, H., Brown, S., Easterling, W., Jallow, B., 2001. Ecosystems
and their goods and services. In: McCarthy, J.J., Canziani, O.F.,
Leary, N.A., Dokken, D.J., White, K.S. (Eds.), Climate Change
2001: Impacts, Adaptation, and Vulnerability. Contribution of
Working Group II to the Third Assessment Report of the
Intergovernmental Panel on Climate Change. Cambridge
University Press, Cambridge, pp. 235–342.
Haxeltine, A., Prentice, I.C., 1996. BIOME3: an equilibrium
terrestrial biosphere model based on ecophysiological constraints, resource availability, and competition among plant
functional types. Glob. Biogeochem. Cy. 10, 693–709.
Hessburg, P., Agee, J.K., 2003. An environmental narrative of
Inland Northwest US forests, 1800–2000. For. Ecol. Manage.
178 (1–2), 23–59.
Heyerdahl, E.K., Brubaker, L.B., Agee, J.K., 1998. Spatial controls
on historical fire regimes: a multiscale example from the
interior West, USA. Ecology 72, 660–679.
Hughes, M.K., Graumlich, L.J., 1996. Climatic Variations and
Forcing Mechanisms of the Last 2000 Years, vol. 141. Multimillennial Dendroclimatic Studies from the Western United
States. NATO ASI Series, pp. 109–124.
Huntley, B., 1995. How vegetation responds to climate change:
evidence from palaeovegetation studies. In: Pernetta, J.C.,
Leemans, R., Elder, D., Humphrey, S. (Eds.), Impacts of
Climate Change on Ecosystems and Species: Environmental
Context. IUCN, Gland, Switzerland, pp. 43–63.
Imbrie, J., Hays, J.D., Martinson, D.G., McIntyre, A., Mix, A.C.,
Morley, J.J., Pisias, N.G., Prell, W.L., Shackleton, N.J., 1984.
The orbital theory of Pleistocene climate: support from a
revised chronology of the marine d18 O record. In: Berger, A.,
Imbrie, J., Kukla, G., Saltzman, B. (Eds.), Milankovitch and
Climate. Reidel, Dordrecht, The Netherlands, pp. 269–305.
IPCC, 2001. Climate change 2001: The scientific basis. Houghton,
J.T., Ding, Y., Griggs, D.J., Noguer, M., van der Linden, P.J.,
Dai, X., Maskell, K., Johnson, C.A. (Eds.), Contribution of
Working Group I to the Third Assessment Report of the
Intergovernmental Panel on Climate Change. Cambridge
University Press, Cambridge, UK, 881 pp.
Iverson, L.R., Prasad, A.M., 1998. Predicting abundance of 80 tree
species following climate change in the eastern United States.
Ecol. Monogr. 68, 465–485.
Keane, R.E., Arno, S.F., Brown, J.K., 1990. Simulating cumulative
fire effects in Ponderosa pine/Douglas-fir forests. Ecology 71,
189–203.
Kirschbaum, M.U.F., Fischlin, A., Cannell, M.G.R., Cruz, R.V.O.,
Galinski, W., Cramer, W.P., 1996. Climate change impacts on
forests. In: Watson, R.T., Zinyowera, M.C., Moss, R.H. (Eds.),
Climate Change 1995: Impacts, Adaptations and Mitigation of
Climate Change. Cambridge University Press, Cambridge, UK,
pp. 95–129.
Knight, D.H., 1987. Parasites, lightning, and the vegetation mosaic
in wilderness landscapes. In: Turner, M.G. (Ed.), Landscape
Heterogeneity and Disturbance. Springer, New York, pp. 59–
83.
Kutzbach, J., Gallimore, R., Harrison, S., Behling, P., Selin, R.,
Laarif, F., 1998. Climate and biome simulations for the past
21,000 years. Quat. Sci. Rev. 17, 473–506.
Leavitt, S.W., 1994. Major wet interval in White Mountains
medieval warm period evidenced in d13 C of bristlecone pine
tree rings. Clim. Change 26, 299–307.
Long, C.J., Whitlock, C., 2002. Fire and vegetation history from
the Coastal rain forest of the Western Oregon Coast Range.
Quat. Res. 58, 215–225.
Long, C.J., Whitlock, C., Bartlein, P.J., Millspaugh, S.H., 1998. A
9000-year fire history from the Oregon Coast Range, based on a
high-resolution charcoal study. Can. J. For. Res. 28, 774–787.
Luckman, B.H., 1994. Evidence for climatic conditions between
ca. 900–1300 A.D. in the southern Canadian Rockies. Clim.
Change 26, 171–182.
Mack, R.N., 1981. Invasion of Bromus tectorum L. into western
North America: an ecological chronicle. Agro-Ecosystems 7,
145–165.
McCreary, D.D., Lavender, D.P., Hermann, R.K., 1990. Predicted
global warming and Douglas-fir chilling requirements. Ann.
Sci. For. 47, 325–330.
Mehringer Jr., P.J., 1985. Late-Quaternary pollen records from the
interior Pacific Northwest and northern Great Basin of the
United States. In: Bryant Jr., V.M., Holloway, R.G. (Eds.),
Pollen Records of Late-Quaternary North American Sediments.
American Association of Stratigraphic Palynologists Foundation, Austin, TX, pp. 167–189.
Mehringer Jr., P.J., Arno, S.F., Peterson, K.L., 1977. Postglacial
history of Lost Trail Pass bog, Bitterroot Mountains, Montana.
Arctic Alpine Res. 9, 345–368.
Meyer, G.A., Pierce, J., 2003. Climatic control of fire-induced
sediment pulses in Yellowstone National Park and Central
Idaho: a long-term perspective. For. Ecol. Manage. 178 (1–2),
89–104.
Meyer, G.A., Wells, S.G., Jull, A.J.T., 1995. Fire and alluvial
chronology in Yellowstone National Park: climatic and intrinsic
controls on Holocene geomorphic processes. Geol. Soc. Am.
Bull. 107, 1211–1230.
Miller, C., Urban, D.L., 1999. Forest pattern, fire, and climatic
change in the Sierra Nevada. Ecosystems 2, 76–87.
Miller, D.A., White, R.A., 1998. A conterminous United States
multilayer soil characteristics dataset for regional climate and
hydrology modeling. Earth Interact. 2, 1–26.
Millspaugh, S.H., Whitlock, C., 1995. A 750-year fire history based
on lake sediment records in central Yellowstone National Park,
USA. Holocene 5, 283–292.
Millspaugh, S.H., Whitlock, C., in press. Postglacial fire, vegetation, and climate history of the Yellowstone-Lamar and Central
Plateau provinces, Yellowstone National Park. In: Wallace, L.
(Ed.), After the Fires: The Ecology of Change in Yellowstone
National Park. Yale University Press, New Haven, CT.
Millspaugh, S.H., Whitlock, C., Bartlein, P.J., 2000. Variations in
fire frequency and climate over the last 17,000 years in central
Yellowstone National Park. Geology 28, 211–214.
Minckley, T., Whitlock, C., 2000. Spatial variation of modern
pollen in Oregon and southern Washington, USA. Rev.
Palaeobot. Palynol. 112, 97–123.
C. Whitlock et al. / Forest Ecology and Management 178 (2003) 5–21
Mitchell, J.F.B., Johns, T.C., 1997. On modification of global
warming by sulphate aerosols. J. Clim. 10, 245–267.
Mock, C.J., 1996. Climatic controls and spatial variations of precipitation in the western United States. J. Clim. 9, 1111–1125.
Mohr, J.A., Whitlock, C., Skinner, C.J., 2000. Postglacial
vegetation and fire history, eastern Klamath Mountains,
California. Holocene 10, 587–601.
Mooney, H.A., Canadell, J., Chapin III, F.S., Ehleringer, J.R.,
Körner, Ch., McMurtrie, R.E., Parton, W.J., Pitelka, L.F.,
Schulze, E.-D., 1999. Ecosystem physiology responses to
global change. In: Walker, B., Steffen, W., Canadell, J.,
Ingram, J. (Eds.), The Terrestrial Biosphere and Global
Change: Implications for Natural and Managed Ecosystems.
Cambridge University Press, Cambridge, UK, pp. 141–189.
National Interagency Fire Center, 2002. http://www.nifc.gov/fireinfo/2000/index.html.
National Research Council, Committee on Ungulate Management
in Yellowstone National Park, 2002. Ecological Dynamics on
Yellowstone’s Northern Range. National Academy Press,
Washington, DC.
National Oceanic and Atmospheric Administration, 2002. http://
www.noaa.gov.
Patterson III, W.A., Edwards, K.J., MacGuire, D.J., 1987.
Microscopic charcoal as a fossil indicator of fire. Quat. Sci.
Rev. 6, 3–23.
Prentice, I.C., Webb III, T., 1998. BIOME6000: reconstructing
global mid-Holocene vegetation patterns from paleoecological
records. J. Biogeogr. 25, 997–1005.
Price, C., Rind, D., 1994. The impact of a 2 CO2 climate on
lightning-caused fires. J. Clim. 7, 1484–1494.
Romme, W.H., Despain, D.G., 1989. Historical perspective on the
Yellowstone fires of 1988. Bioscience 39, 695–698.
Romme, W.H., Turner, M.G., 1991. Implications of global climate
change for biogeographic patterns in the Greater Yellowstone
ecosystem. Conserv. Biol. 5, 373–386.
Shafer, S.L., 2000. Potential vegetation response to future climate
change in western North America and its implications for
biological conservation and geographical conceptualizations of
place. Ph.D. Dissertation. University of Oregon, Eugene, OR.
Shafer, S.L., Bartlein, P.J., Thompson, R.S., 2001. Potential changes
in the distributions of western North America tree and shrub taxa
under future climate scenarios. Ecosystems 4, 200–215.
Simberloff, D., 2000. Global climate change and introduced
species in United States forests. Sci. Total Environ. 262, 253–
261.
Sitch, S., Smith, B., Prentice, I.C., Arneth, A., Bondeau, A.,
Cramer, W., Kaplan, J.O., Levis, S., Lucht, W., Sykes, M.T.,
Thonicke, K., Venevsky, S., 2003. Evaluation of ecosystem
dynamics, plant geography and terrestrial carbon cycling in the
LPJ dynamic global vegetation model. Global Change Biology
9, 161–185.
21
Stine, S., 1994. Extreme and persistent drought in California and
Patagonia during medieval time. Nature 269, 546–549.
Swetnam, T., Allen, C., Betancourt, J., 1999. Applied historical
ecology: using the past to manage for the future. Ecol. Appl. 9,
1189–1206.
Thompson, R.S., Whitlock, C., Bartlein, P.J., Harrison, S.P.,
Spaulding, W.G., 1993. Climatic changes in the western United
States since 18,000 year BP. In: Wright Jr., H.E., Kutzbach,
J.E., Webb III, T., Ruddiman, W.F., Street-Perrott, F.A. (Eds.),
Global Climates Since the Last Glacial Maximum. University
of Minnesota Press, Minneapolis, MN, pp. 468–513.
Thompson, R.S., Anderson, K.H., Bartlein, P.J., 1999. Atlas of
relations between climatic parameters and distributions of
important trees and shrubs in North America. US Geological
Survey Professional Paper No. 1650 A&B.
USDA
Forest
Service,
2001.
http://www.fireplan.gov/
FS_Prog_RPT_July_30.cfm.
Webb III, T., 1988. Eastern North America. In: Huntley, B., Webb
III, T. (Eds.), Vegetation History. Kluwer Academic Publishers,
Dordrecht, The Netherlands, pp. 385–414.
Webb III, T., 1995. Pollen records of late Quaternary vegetation
change: plant community rearrangements and evolutionary
implications. In: National Research Council Commission on
Geosciences, Environment, and Resources. Effects of Past
Global Change on Life. National Academy Press, Washington,
DC, pp. 221–232.
Weisberg, P.J., Swanson, F.J., 2003. Regional synchroneity in fire
regimes of western Oregon and Washington, USA. For. Ecol.
Manage. 172, 17–28.
West, N.E., Young, J.A., 2000. Intermountain valleys and lower
mountain slopes. In: Barbour, M.G., Billings, W.D. (Eds.),
North American Terrestrial Vegetation. Cambridge University
Press, Cambridge, UK, pp. 255–284.
Whitlock, C., 1992. Vegetational and climatic history of the Pacific
Northwest during the last 20,000 years: implications for
understanding present-day biodiversity. Northw. Environ. J. 8,
5–28.
Whitlock, C., 1993. Postglacial vegetation and climate of Grand
Teton and southern Yellowstone National Parks. Ecol. Monogr.
63, 173–198.
Whitlock, C., Bartlein, P.J., 1993. Spatial variations of Holocene
climatic change in the Yellowstone region. Quat. Res. 39, 231–
238.
Whitlock, C., Larsen, C.P.S., 2002. Charcoal as a fire proxy. In:
Smol, J.P., Birks, H.J.B., Last, W.M. (Eds.), Tracking
Environmental Change Using Lake Sediments, vol. 3: Terrestrial, Algal, and Siliceous Indicators. Kluwer Academic
Publishers, Dordrecht, The Netherlands, pp. 74–97.
Whitlock, C., Millspaugh, S.H., 1996. Testing assumptions of fire
history studies: an examination of modern charcoal accumulation in Yellowstone National Park. Holocene 6, 7–15.